WO2025169329A1 - Information processing device, information processing method, and recording medium - Google Patents

Abstract

An information processing device according to the present invention comprises: a detection means that uses a computation model to detect, for a search target image to be searched and a plurality of registered pattern images which have been registered, singularities included in the respective images; and a selection means that, on the basis of the singularities, selects an image of interest which is to be compared to the search target image from among the plurality of registered pattern images, wherein the computation model, upon receipt of input of a pattern image, outputs singularity information that indicates a singularity included in the pattern image.

Description

Information processing device, information processing method, and recording medium

This disclosure relates to the technical fields of information processing devices, information processing methods, and recording media.

It is known that features extracted from palm print images are used for palm print matching. For example, Patent Document 1 discloses palm print matching by extracting features from pre-registered palm print images and from palm print images acquired during matching.

International Publication No. 2023/281563

This disclosure aims to improve the related technologies mentioned above.

One aspect of the information processing device disclosed herein includes a detection means that uses a computational model to detect singular points contained in a search target image and each of a plurality of registered pattern images, and a selection means that selects a target image to be matched with the search target image from the plurality of registered pattern images based on the singular points; when a pattern image is input, the computational model outputs singular point information indicating the singular points contained in the pattern image.

One aspect of the information processing method disclosed herein is an information processing method that uses a computational model executed by a computer to detect singular points contained in a search target image and each of a plurality of registered pattern images, and selects a target image to be matched with the search target image from the plurality of registered pattern images based on the singular points. When a pattern image is input, the computational model outputs singular point information indicating the singular points contained in the pattern image.

One aspect of the recording medium disclosed herein is an information processing method that uses a computational model to detect singular points contained in a search target image and each of a plurality of registered pattern images, and selects a target image from the plurality of registered pattern images to be matched with the search target image based on the singular points. The computational model has recorded thereon a computer program that causes a computer to execute the information processing method, which, when a pattern image is input, outputs singular point information indicating the singular points contained in the pattern image.

1 is a block diagram showing a configuration of an information processing device according to the present disclosure. 1 is a flowchart showing a flow of information processing operations in an information processing device according to the present disclosure. 1 is a block diagram showing a configuration of an information processing device according to the present disclosure. 1 is a flowchart showing a flow of information processing operations in an information processing device according to the present disclosure. FIG. 1 is a schematic diagram illustrating an outline of a singularity. 1 is a schematic diagram illustrating an information processing method in an information processing device according to the present disclosure. 1 is a block diagram showing a configuration of an information processing device according to the present disclosure. 1 is a schematic diagram illustrating an information processing method in an information processing device according to the present disclosure. 1 is a block diagram showing a configuration of an information processing device according to the present disclosure. 1 is a schematic diagram illustrating an information processing method in an information processing device according to the present disclosure.

Hereinafter, embodiments of an information processing device, an information processing method, and a recording medium will be described with reference to the drawings.
[1: First embodiment]

A first embodiment of an information processing device, an information processing method, and a recording medium will be described. Hereinafter, a first embodiment of an information processing device, an information processing method, and a recording medium will be described using an information processing device 1 according to this disclosure.
[1-1: Configuration of information processing device 1]

The configuration of the information processing device 1 according to this disclosure will be described with reference to Figure 1. Figure 1 is a block diagram showing the configuration of the information processing device 1 according to this disclosure.

As shown in FIG. 1, the information processing device 1 includes a calculation device 11, a storage device 12, and a communication device 13. The calculation device 11, the storage device 12, and the communication device 13 may be connected via a data bus 16.

The computing device 11 includes at least one processor (i.e., one processor or multiple processors) as hardware. The processor may include, for example, a processor conforming to a von Neumann computer architecture. A processor conforming to a von Neumann computer architecture may include at least one of a CPU (Central Processing Unit) and a GPU (Graphics Processing Unit). The processor may include, for example, a processor conforming to a non-von Neumann computer architecture. A processor conforming to a non-von Neumann computer architecture may include at least one of an FPGA (Field Programmable Gate Array) and an ASIC (Application Specific Circuit).

The arithmetic device 11 reads a computer program 121 including at least one of computer program code and computer program instructions. For example, the arithmetic device 11 may read a computer program 121 stored in the storage device 12. For example, the arithmetic device 11 may read a computer program 121 stored in a computer-readable, non-transitory storage medium using a storage medium reading device (not shown) provided in the information processing device 1. The computer program 121 read from the storage medium may be stored in the storage device 12. The arithmetic device 11 may acquire (i.e., download or read) the computer program 121 from a device (not shown) located outside the information processing device 1 via the communication device 13 (or another communication device). The downloaded computer program 121 may be stored in the storage device 12.

The arithmetic device 11 executes the loaded computer program 121. As a result, logical functional blocks for executing the processing to be performed by the information processing device 1 (e.g., the information processing described below) are realized within the arithmetic device 11. In other words, the arithmetic device 11, together with the storage device 12, etc. in which the computer program 121 is recorded (in other words, together with the storage device 12 and the computer program 121 recorded in the storage device 12, etc.), can function as a controller or computer for realizing the logical functional blocks for executing the processing to be performed by the information processing device 1. In other words, the at least one processor provided in the arithmetic device 11, the memory (recording medium) provided in the storage device 12, etc., and the computer program 121 are configured so that the information processing device 1 performs the processing to be performed by the information processing device 1 (e.g., the information processing described below). The arithmetic device 11 may output information to another computer, cloud server, or other device (not shown) located outside the information processing device 1 via the communication device 13 (or other communication device).

The recording medium for recording the computer program 121 executed by the arithmetic device 11 may be at least one of the following: a CD-ROM, CD-R, CD-RW, flexible disk, MO, DVD-ROM, DVD-RAM, DVD-R, DVD+R, DVD-RW, DVD+RW, and optical disks such as Blu-ray (registered trademark), magnetic media such as magnetic tape, magneto-optical disk, semiconductor memory such as USB memory, and any other medium capable of storing a program. The recording medium may include a device capable of recording the computer program 121 (for example, a general-purpose device or dedicated device in which the computer program 121 is implemented in a state in which it can be executed in at least one of the forms of software and firmware). Furthermore, each process or function included in the computer program 121 may be realized by a logical processing block realized within the arithmetic device 11 when the arithmetic device 11 (i.e., processor) executes the computer program 121, or may be realized by hardware such as a predetermined gate array (FPGA (Field Programmable Gate Array), ASIC (Application Specific Integrated Circuit)) included in the arithmetic device 11, or may be realized in a form that combines logical processing blocks and partial hardware modules that realize some of the hardware elements.

A computational model M that can be constructed by machine learning is implemented within the computational device 11 by the computational device 11 executing the computer program 121. An example of a computational model M that can be constructed by machine learning is a computational model M that includes a neural network (so-called artificial intelligence (AI)). In this case, learning of the computational model M may include learning of neural network parameters (e.g., at least one of weights and biases). The computational device 11 executes at least the detection process using the computational model M. A computational model M that has been constructed by machine learning may be implemented in the computational device 11. A computational model M that has been constructed by offline machine learning using training data may be implemented in the computational device 11. Furthermore, the computational model M implemented in the computational device 11 may be updated by online machine learning on the computational device 11. Alternatively, the calculation device 11 may execute information processing using a calculation model M implemented in a device external to the calculation device 11 (i.e., a device provided external to the information processing device 1) in addition to or instead of the calculation model M implemented in the calculation device 11.

FIG. 1 shows an example of logical functional blocks implemented within the calculation device 11 to perform information processing. As shown in FIG. 1, a detection unit 111 and a selection unit 112 are implemented within the calculation device 11. The detection unit 111 executes the process of detecting singular points using the calculation model M described above. The processes performed by the detection unit 111 and the selection unit 112 will be explained with reference to FIG. 2.

The storage device 12 includes at least one memory capable of storing desired data. In other words, the storage device 12 includes at least one memory containing desired data. For example, the storage device 12 may store a computer program 121 executed by the arithmetic device 11. In this case, the storage device 12 (memory) may be used as the above-mentioned recording medium for recording the computer program 121 executed by the arithmetic device 11. The storage device 12 may temporarily store data used by the arithmetic device 11 when the arithmetic device 11 is executing the computer program 121. The storage device 12 may also store data that the information processing device 1 will store long-term. The storage device 12 may include at least one of RAM (Random Access Memory), ROM (Read Only Memory), a hard disk device, a magneto-optical disk device, an SSD (Solid State Drive), and a disk array device. In other words, the storage device 12 may include a non-temporary recording medium.

The communication device 13 is capable of communicating with devices external to the information processing device 1 and the information processing device 2 via a communication network (not shown). The communication device 13 may be a communication interface based on standards such as Ethernet (registered trademark), Wi-Fi (registered trademark), Bluetooth (registered trademark), or USB (Universal Serial Bus).
[1-2: Information Processing Method Executed by Information Processing Device 1]

The information processing method executed by information processing device 1 will be described with reference to Figure 2. Figure 2 is a flowchart showing an example of the flow of the information processing method executed by information processing device 1.

As shown in FIG. 2, the detection unit 111 uses the computational model M to detect singular points S contained in the image for the search target image Q to be searched and for each of the multiple registered pattern images (step S11). When a pattern image is input, the computational model M outputs singular point information indicating the singular points S contained in the pattern image. The detection unit 111 detects singular points contained in the search target image Q (sometimes referred to as "query singular points") and singular points contained in each of the multiple registered registered pattern images (sometimes referred to as "target singular points").

The selection unit 112 selects a target image to be matched with the search target image from among the multiple registered pattern images based on the singular points (step S12). In other words, the selection unit 112 selects a target to be matched with the search target image from among the multiple registered pattern images based on the singular points. Note that the query singular points and the target singular points may be detected at different times. For example, the target singular points may be detected when the registered pattern image is registered and registered together with the registered pattern image.
[1-3: Technical Effects of Information Processing Device 1]

The information processing device 1 according to this disclosure selects a target image to be matched with the search target image Q in accordance with the search target image Q. In other words, the information processing device 1 limits the targets for matching, thereby reducing the processing load for matching.
[2: Second embodiment]

A second embodiment of an information processing device, an information processing method, and a recording medium will be described. Below, a second embodiment of an information processing device, an information processing method, and a recording medium will be described using an information processing device 2 disclosed herein.

In the second embodiment, a case where this disclosure is applied to a palm print image, which is a pattern image, is described. Note that this disclosure can be applied to a relatively wide range of pattern images that may include multiple singular points S, such as images of patterns on the soles of feet, in addition to palm print images.
[2-1: Configuration of information processing device 2]

The configuration of the information processing device 2 disclosed herein will be described with reference to Figure 3. Figure 3 is a block diagram showing the configuration of the information processing device 2 disclosed herein.

As shown in FIG. 3, the information processing device 2 may further include an input device 14 and an output device 15 in addition to the calculation device 11, the storage device 12, and the communication device 13. However, the information processing device 2 does not have to include at least one of the input device 14 and the output device 15. The calculation device 11, the storage device 12, the communication device 13, the input device 14, and the output device 15 may be connected via a data bus 16.

As shown in FIG. 3, in addition to a detection unit 211 and a selection unit 212, a reception unit 213, an extraction unit 214, and a matching unit 215 are further implemented within the calculation device 11 in the second embodiment. Furthermore, a registered palm print image database DB may be implemented within the storage device 12 in the second embodiment. The registered palm print image database DB may also be implemented in a storage device external to the information processing device 2.

The input device 14 is a device that accepts information input to the information processing device 2 from outside the information processing device 2. For example, the input device 14 may include an operating device (e.g., at least one of a keyboard, mouse, and touch panel) that can be operated by an operator of the information processing device 2. For example, the input device 14 may include a reading device that can read information recorded as data on a recording medium that can be attached externally to the information processing device 2.

The output device 15 is a device that outputs information to the outside of the information processing device 2. For example, the output device 15 may output information as an image. That is, the output device 15 may include a display device (a so-called display) that can display an image showing the information to be output. For example, the output device 15 may output information as sound. That is, the output device 15 may include an audio device (a so-called speaker) that can output sound. For example, the output device 15 may output information on paper. That is, the output device 15 may include a printing device (a so-called printer) that can print desired information on paper.

The information processing device 2 is configured as a device for selecting a target image, similar to the information processing device 1. Furthermore, the information processing device 2 is configured as a device for matching palm prints. The information processing device 2 matches palm prints using the selected target image.
[2-2: Information Processing Method Executed by Information Processing Device 2]

The information processing operation in the information processing device 2 will be described with reference to Figure 4. Figure 4 is a flowchart showing the flow of the information processing operation in the information processing device 2.

As shown in FIG. 4, the reception unit 213 receives a request to match a search target image Q (step S21). The request to match a search target image Q is a request to match the search target image Q with a registered palm print image R registered in the registered palm print image database DB. FIG. 5(a) shows an example of a search target image Q and a registered palm print image R.

Registered palm print image R is an image of an imprinted palm print. It is preferable that registered palm print image R is an image of the entire palm print. In contrast, search target image Q may be an image of the entire palm print, or an image of a portion of the palm print. Search target image Q may be an image of an imprinted palm print, or an image of a left-behind palm print. Note that an imprinted palm print is a palm print obtained by imprinting.

The detection unit 211 uses the computational model M to detect singular points S contained in the search target image Q and each of the multiple registered palm print images R (step S22). Figure 5(b) shows an example of the detected query singular points QS and target singular points RS.

When a pattern image is input, the computational model M outputs singularity information indicating the singularity S contained in the pattern image. As illustrated in FIG. 5, when a search target image Q is input, the computational model M outputs singularity information indicating the query singularity QS contained in the search target image Q. Furthermore, when a registered palmprint image R is input, the computational model M outputs singularity information indicating the target singularity RS contained in the registered palmprint image R. The singularity information may indicate at least one of the type of singularity S, the position of the singularity S, and the direction of the singularity S.

The computational model M is constructed by machine learning. The computational model M may be constructed by machine learning using training data. The training data may be data in which a pattern image is accompanied by correct answer information indicating at least one of the type of singular point contained in the pattern image, the position of the singular point, and the direction of the singular point. In this case, the computational model M is trained so that when a pattern image is input, it can output correct answer information. The computational model M is constructed so that when a pattern image is input, it can infer at least one of the type of singular point contained in the pattern image, the position of the singular point, and the direction of the singular point.

Types of singular points S include at least cores (which may be referred to as "centers") and deltas (which may be referred to as "deltas"). As shown in Figure 6(a), a point where the direction of the palm print ridges changes suddenly is called a core. As shown in Figure 6(b), a point where a delta is formed by the palm print ridges is called a delta.

For example, as shown in FIG. 6(a), the computational model M may extract the center Sc of the looped ridge as the position of the core. Furthermore, as shown in FIG. 6(b), the computational model M may extract a position Sd within a delta formed by the ridge as the position of the delta. The position of the singular point S may be represented by coordinates in the pattern image. The position of the singular point S may also be a relative position in the palm print region.

The direction of the singularity point S may be determined according to the direction of the ridges surrounding the singularity point S. For example, the computational model M may infer the direction of the singularity point S as shown by the arrow in Figure 5(a). The computational model M may also infer the direction of the singularity point S as shown by the arrow in Figure 5(b). In other words, when the type of the singularity point S is delta, the computational model M may infer three directions.

Based on the singularity information, the detection unit 211 performs at least one of the following: identifying the type of the detected singularity, identifying the position of the singularity, and identifying the direction of the singularity.

Note that the detection unit 211 may detect the query singularity and the target singularity at different times. For example, the detection unit 211 may detect the target singularity at the time when the registered palm print image R is registered in the registered palm print image database DB. In this case, information indicating the target singularity may be registered in association with the registered palm print image R in the registered palm print image database DB.

The selection unit 212 selects a target image to be matched with the search target image Q from the multiple registered palm print images R based on the singular points S (step S23). The selection unit 212 may select multiple target images. In other words, the selection unit 212 excludes from the multiple registered palm print images R those that do not need to be used as targets for matching. In other words, the selection unit 212 retains from the multiple registered palm print images R those that need to be used as targets for matching.

In other words, the selection unit 212 reduces the targets of processing from step S24 onwards. In other words, the selection unit 212 narrows down the targets of processing from step S24 onwards. In other words, the information processing device 2 performs filtering processing before feature point matching.

The extraction unit 214 extracts feature points from the search target image Q and each of the target images (step S24). The extraction unit 214 may extract the end points and branch points of palm print ridges as feature points.

The matching unit 215 matches the search target image Q with the target image based on the feature points (step S25). For example, the matching unit 215 may find a similarity between the positional relationship of the feature points included in the search target image Q and the positional relationship of the feature points included in the target image, and match the search target image Q with the target image.
[2-3: Technical Effects of Information Processing Device 2]

Compared to fingerprints, palm prints have a larger area to be matched. For this reason, palm print matching often requires higher computational costs than fingerprint matching.

The information processing device 2 according to this disclosure detects singularities and reduces the number of matching targets, thereby reducing the computational cost for matching compared to when the number of matching targets is not reduced. Furthermore, the information processing device 2 can detect singularities that are useful for selecting target images using a computational model constructed by machine learning.
[3: Third embodiment]

A third embodiment of an information processing device, an information processing method, and a recording medium will be described. Hereinafter, a third embodiment of an information processing device, an information processing method, and a recording medium will be described using an information processing device 3 according to this disclosure.
[3-1: Information processing method executed by information processing device 3]

Information processing device 3, like information processing device 1 and information processing device 2, is configured as a device for selecting a target image. Furthermore, like information processing device 2, information processing device 3 may be configured as a device for matching palm prints. The third embodiment differs from the first and second embodiments in the operation of the selection unit 312.

In the third embodiment, the selection unit 312 selects a target image based on a comparison between the positional relationship of singular points S included in the search target image Q and the positional relationship of singular points S included in the registered palm print image R. As illustrated in FIG. 8, the selection unit 312 may designate the singular point S located closest to the ball of the foot as A, and the adjacent singular points S in a counterclockwise direction as B, C, and D, and compare the positional relationship of the singular points S. FIG. 8(a) shows an example of a query singular point QS, and FIG. 8(b) shows an example of a target singular point RS.

For example, the selection unit 312 may select a target image based on the similarity between the positional relationship of the query singular points QS included in the search target image Q and the positional relationship of the target singular points RS included in the registered palmprint image R. If the similarity between the positional relationship of the query singular points QS included in the search target image Q and the positional relationship of the target singular points RS included in the registered palmprint image R is greater than a criterion, the selection unit 312 may select the registered palmprint image R as the target image, and if it is less than the criterion, exclude the registered palmprint image R from the target images.

Specifically, the selection unit 312 may select a target image based on at least a portion of the difference in distance between two corresponding singular points S between the search target image Q and the registered palm print image R, and the difference in each of the angles of a triangle formed by three points. FIG. 8(c) shows an example of the distance between two points and the angle of a triangle formed by three points in the search target image Q, and FIG. 8(d) shows an example of the distance between two points and the angle of a triangle formed by three points in the registered palm print image R. For example, the selection unit 312 may select as the target image a registered palm print image R in which the difference in distance between corresponding singular points S is less than a predetermined value and the angle formed by the corresponding singular points S is less than a predetermined value. The selection unit 312 may also select as the target image a registered palm print image R in which the difference in x coordinates of corresponding singular points S is less than a predetermined value, the difference in y coordinates of corresponding singular points S is less than a predetermined value, and the angle formed by the corresponding singular points S is less than a predetermined value. Alternatively, if the difference in distance between corresponding singular points S is equal to or greater than a predetermined value, or if the angle formed by corresponding singular points S is equal to or greater than a predetermined value, the selection unit 312 may exclude the corresponding registered palm print image R from being subjected to matching.

Alternatively, the selection unit 312 may calculate the difference in distance between two points and the difference in each angle of a triangle formed by three points for all of the singular points S included in the search target image Q, accumulate the differences, and select as the target image a registered palm print image R for which the accumulated difference is less than a predetermined value. The selection unit 312 only needs to be able to exclude from the matching target a registered palm print image R whose positional relationship of the singular points S is clearly different from that of the search target image Q.

Furthermore, the selection unit 312 may use the direction of the singular point S. For example, the selection unit 312 may add a restriction to the distance and angle restrictions, such as whether the difference in direction of the corresponding singular point S between the search target image Q and the registered palm print image R is less than a predetermined value, and select the registered palm print image R as the target image.

The selection unit 312 may also compare the types of singular points present at corresponding positions between the search target image Q and the registered palm print image R. For example, the selection unit 312 may determine that singular points S do not correspond even if they exist at corresponding positions between the search target image Q and the registered palm print image R, if the types are different (for example, one is a core and the other is a delta).

It should be noted that the distance between two target singular points RS in the registered palm print image R and the angle of a triangle formed by three target singular points RS in the registered palm print image R may be determined in advance. In this case, information indicating the distance and angle may be associated with the registered palm print image R and registered in the registered palm print image database DB.
[3-2: Technical Effects of Information Processing Device 3]

The information processing device 3 disclosed herein excludes registered palm print images R that are not similar overall to the search target image Q from the matching target, thereby reducing the processing load for matching while maintaining matching accuracy.
[4: Fourth embodiment]

Fourth Embodiment of Information Processing Apparatus, Information Processing Method, and Recording Medium Will Be Described Hereinafter, a fourth embodiment of an information processing apparatus, an information processing method, and a recording medium will be described using an information processing apparatus 4 according to this disclosure.
[4-1: Information processing method executed by information processing device 4]

Information processing device 4 is configured as a device for selecting a target image, similar to information processing device 1 to information processing device 3. Furthermore, information processing device 4 may be configured as a device for matching palm prints, similar to information processing device 2 and information processing device 3. The fourth embodiment differs from the first to third embodiments in the operation of the selection unit 412.

In the fourth embodiment, the selection unit 412 selects a target image based on the type of any singular point included in the search target image Q. The selection unit 412 may select a target image based on the type and direction of any singular point included in the search target image Q. The selection unit 412 may select a target image based on the type and direction of one singular point included in the search target image Q.
[a: When the search target image Q is an image of a palm print portion]

In the fourth embodiment, the search target image Q may be an image including part of a palm print, such as an image of a palm print that has been left behind. Figure 10 illustrates an example of the outline of the operation of the selection unit 412. Figure 10(a) illustrates an example of a search target image Q in the fourth embodiment. The search target image Q illustrated in Figure 10(a) includes one core S.

FIGS. 10(b) and 10(c) show examples of registered palm print images R. The registered palm print image Rb shown in FIG. 10(b) includes one core b and four deltas. The registered palm print image Rc shown in FIG. 10(c) includes two cores c1 and c2 and two deltas.

In the case illustrated in FIG. 10(b), the selection unit 412 in the fourth embodiment may select, as the target image, a predetermined peripheral area Ab of core b in registered palm print image Rb. Furthermore, in the case illustrated in FIG. 10(c), the selection unit 412 in the fourth embodiment may select, as the target image, each of a predetermined peripheral area Ac1 of core c1 in registered palm print image Rc and a predetermined peripheral area Ac2 of core c2 in registered palm print image Rc. In other words, the selection unit 412 in the fourth embodiment selects, as the target image, a partial area of registered palm print image R. In other words, the selection unit 412 limits the area to be matched.

Also, as illustrated in FIG. 10, the selection unit 412 may select a predetermined peripheral area of the singular point S as the target image, in accordance with the direction of the singular point S of the search target image Q.

The predetermined surrounding area A may be an area that may be associated with the singular point S. The predetermined surrounding area A may be an area that may be influenced by the singular point S.
[b: When the search target image Q is an image of the entire palm print]

Alternatively, the selection unit 412 may cut out a predetermined area surrounding the singular point S from the search target image Q, which is an image of the entire palm print, as shown in FIG. 8(a), and then cut out a partial area of the registered palm print image R based on the type and direction of the singular point S, and select it as the target image.

In the fourth embodiment, the selection operation by the selection unit 312 in the third embodiment may also be performed. For example, a registered palm print image R may be selected based on a comparison of the positional relationships of the singular points S, and a partial region may be cut out from the selected registered palm print image R and selected as the target image.
[4-2: Technical Effects of Information Processing Device 4]

When the search target image Q is an image of a portion of a palm print, the information processing device 4 according to this disclosure can select, based on singular points, partial regions that are likely to have similar features. Furthermore, even when the search target image Q is an image of the entire palm print, the information processing device 4 can select, for example, partial regions that are useful for matching. In this way, the information processing device 4 can maintain matching accuracy while reducing the processing load of matching.
[5: Supplementary Note]

The above-described embodiment may be further described as follows, but is not limited to the following.
[Appendix 1]
a detection means for detecting, using a computational model, singular points included in an image for a search target image and each of a plurality of registered pattern images;
a selection means for selecting a target image to be matched with the search target image from the plurality of registered pattern images based on the singular points,
When a pattern image is input, the computational model outputs singular point information indicating singular points included in the pattern image.
[Appendix 2]
The information processing device according to claim 1, wherein the computational model is constructed based on machine learning.
[Appendix 3]
The information processing device according to claim 1, wherein the pattern image is a palm print image.
[Appendix 4]
The information processing apparatus according to claim 1 , wherein when the pattern image is input, the computational model infers at least one of the type of the singular point included in the pattern image, the position of the singular point, and the direction of the singular point.
[Appendix 5]
The information processing device according to claim 1, wherein the singularity information indicates at least one of a type of the singularity, a position of the singularity, and a direction of the singularity.
[Appendix 6]
The information processing device according to claim 5, wherein the detection means performs at least one of identifying a type of the singular point, identifying a position of the singular point, and identifying a direction of the singular point based on the singular point information.
[Appendix 7]
The information processing device according to claim 1, wherein the selection means selects the target image based on a comparison between a positional relationship of a singular point included in the search target image and a positional relationship of a singular point included in the registered pattern image.
[Appendix 8]
The information processing device according to claim 1, wherein the selection means selects the target image based on a degree of similarity between a positional relationship of a singular point included in the search target image and a positional relationship of a singular point included in the registered pattern image.
[Appendix 9]
The information processing device according to Supplementary Note 1, wherein the selection means selects the target image based on at least a part of a difference in distance between two corresponding singular points between the image to be searched and the registered pattern image, and a difference in each angle of a triangle formed by three corresponding points between the image to be searched and the registered pattern image.
[Supplementary Note 10]
The information processing device according to claim 1, wherein the selection means selects the target image based on a comparison of types of corresponding singular points between the search target image and the registered pattern image.
[Appendix 11]
The information processing device according to claim 1, wherein the selection means selects a partial region of the registered pattern image based on a type of any singular point included in the image to be searched.
[Appendix 12]
The information processing device according to claim 1, wherein the selection means selects a partial region of the registered pattern image based on a type and a direction of any singular point included in the image to be searched.
[Appendix 13]
extraction means for extracting feature points from the search target image and each of the target images;
The information processing apparatus according to claim 1 , further comprising: a matching unit that matches the search target image with the target image based on the feature points.
[Appendix 14]
The computer executes
Using the computational model, a singular point contained in the image is detected for the search target image and each of the plurality of registered pattern images;
selecting a target image to be matched with the search target image from the plurality of registered pattern images based on the singular points,
When a pattern image is input, the computational model outputs singular point information indicating singular points included in the pattern image.
[Appendix 15]
Using the computational model, a singular point contained in the image is detected for the search target image and each of the plurality of registered pattern images;
selecting a target image to be matched with the search target image from the plurality of registered pattern images based on the singular points,
The computational model outputs singular point information indicating singular points contained in the pattern image when a pattern image is input.

This disclosure may be modified as appropriate within the scope of the claims and the spirit or concept of the invention as can be read from the entire specification, and information processing devices, information processing methods, and recording media incorporating such modifications are also included within the technical concept of this disclosure.

1, 2, 3, 4 Information processing device 111, 211 Detection unit 112, 212, 312, 412 Selection unit 213 Reception unit 214 Extraction unit 215 Matching unit DB Registered palm print image database Q Search target image R Registered palm print image S Singular point QS Query singular point RS Target singular point

Claims (15)

a detection means for detecting, using a computational model, singular points included in an image for a search target image and each of a plurality of registered pattern images;
a selection means for selecting a target image to be matched with the search target image from the plurality of registered pattern images based on the singular points,
When a pattern image is input, the computational model outputs singular point information indicating singular points included in the pattern image. The information processing device according to claim 1 , wherein the computational model is constructed based on machine learning. The information processing device according to claim 1 , wherein the pattern image is a palm print image. The information processing apparatus according to claim 1 , wherein when the pattern image is input, the computational model infers at least one of the type of the singular point included in the pattern image, the position of the singular point, and the direction of the singular point. The information processing apparatus according to claim 1 , wherein the singularity information indicates at least one of a type of the singularity, a position of the singularity, and a direction of the singularity. The information processing apparatus according to claim 5 , wherein the detection means performs at least one of identifying a type of the singular point, identifying a position of the singular point, and identifying a direction of the singular point based on the singular point information. The information processing apparatus according to claim 1 , wherein the selection means selects the target image based on a comparison between a positional relationship of specific points included in the search target image and a positional relationship of specific points included in the registered pattern image. The information processing apparatus according to claim 1 , wherein the selection means selects the target image based on a similarity between a positional relationship of specific points included in the search target image and a positional relationship of specific points included in the registered pattern image. 2. The information processing device according to claim 1, wherein the selection means selects the target image based on at least a part of a difference in distance between two corresponding singular points between the search target image and the registered pattern image, and a difference in each angle of a triangle formed by three corresponding points between the search target image and the registered pattern image. The information processing apparatus according to claim 1 , wherein the selection means selects the target image based on a comparison of types of corresponding singular points between the search target image and the registered pattern image. The information processing apparatus according to claim 1 , wherein the selection means selects a partial region of the registered pattern image based on a type of any singular point included in the image to be searched. The information processing apparatus according to claim 1 , wherein the selection means selects a partial region of the registered pattern image based on the type and direction of any singular point included in the image to be searched. extraction means for extracting feature points from the search target image and each of the target images;
The information processing apparatus according to claim 1 , further comprising: a matching unit that matches the search target image with the target image based on the feature points. The computer executes
Using the computational model, a singular point contained in the image is detected for the search target image and each of the plurality of registered pattern images;
selecting a target image to be matched with the search target image from the plurality of registered pattern images based on the singular points,
When a pattern image is input, the computational model outputs singular point information indicating singular points included in the pattern image. Using the computational model, a singular point contained in the image is detected for the search target image and each of the plurality of registered pattern images;
selecting a target image to be matched with the search target image from the plurality of registered pattern images based on the singular points,
The computational model outputs singular point information indicating singular points contained in the pattern image when the pattern image is input. A recording medium having a computer program recorded thereon that causes a computer to execute an information processing method.